social-card-gen
Pass
Audited by Gen Agent Trust Hub on Apr 4, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The
generate.jsscript includes functionality to fetch content from remote URLs using thefetchAPI when the--urlflag is used. - [COMMAND_EXECUTION]: The skill includes a standalone Node.js script (
generate.js) and provides instructions for users to execute it locally to automate post generation. - [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it processes untrusted text from external files or URLs without sufficient sanitization or boundary markers.
- Ingestion points: Untrusted data enters the agent's context through file reads (
--file) and network fetches (--url) ingenerate.js, or via direct text input in the manual path. - Boundary markers: The skill does not employ delimiters or instructions to ignore embedded commands within the source content.
- Capability inventory: The tool can read from the filesystem, write to the filesystem, and perform network operations.
- Sanitization: Input cleaning is limited to removing Markdown and HTML tags, which does not prevent adversarial natural language instructions from influencing the AI's output.
Audit Metadata